[nrktv:season,series] Fix extraction and update tests (closes #17159, closes #17258)

1 files changed, 43 insertions, 25 deletions

diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py
index c5001ef48..48bc6fd7a 100644
--- a/youtube_dl/extractor/nrk.py
+++ b/youtube_dl/extractor/nrk.py
@@ -211,13 +211,13 @@ class NRKIE(NRKBaseIE):
     _TESTS = [{
         # video
         'url': 'http://www.nrk.no/video/PS*150533',
-        'md5': '2f7f6eeb2aacdd99885f355428715cfa',
+        'md5': '706f34cdf1322577589e369e522b50ef',
         'info_dict': {
             'id': '150533',
             'ext': 'mp4',
             'title': 'Dompap og andre fugler i Piip-Show',
             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
-            'duration': 263,
+            'duration': 262,
         }
     }, {
         # audio
@@ -256,14 +256,14 @@ class NRKTVIE(NRKBaseIE):
     _API_HOSTS = ('psapi-ne.nrk.no', 'psapi-we.nrk.no')
     _TESTS = [{
         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
-        'md5': '4e9ca6629f09e588ed240fb11619922a',
+        'md5': '9a167e54d04671eb6317a37b7bc8a280',
         'info_dict': {
             'id': 'MUHH48000314AA',
             'ext': 'mp4',
             'title': '20 spørsmål 23.05.2014',
             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
             'duration': 1741,
-            'series': '20 spørsmål - TV',
+            'series': '20 spørsmål',
             'episode': '23.05.2014',
         },
     }, {
@@ -301,7 +301,7 @@ class NRKTVIE(NRKBaseIE):
                 'id': 'MSPO40010515AH',
                 'ext': 'mp4',
                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 1)',
-                'description': 'md5:c03aba1e917561eface5214020551b7a',
+                'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
                 'duration': 772,
                 'series': 'Tour de Ski',
                 'episode': '06.01.2015',
@@ -314,7 +314,7 @@ class NRKTVIE(NRKBaseIE):
                 'id': 'MSPO40010515BH',
                 'ext': 'mp4',
                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 2)',
-                'description': 'md5:c03aba1e917561eface5214020551b7a',
+                'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
                 'duration': 6175,
                 'series': 'Tour de Ski',
                 'episode': '06.01.2015',
@@ -326,7 +326,7 @@ class NRKTVIE(NRKBaseIE):
         'info_dict': {
             'id': 'MSPO40010515',
             'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015',
-            'description': 'md5:c03aba1e917561eface5214020551b7a',
+            'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
         },
         'expected_warnings': ['Video is geo restricted'],
     }, {
@@ -406,21 +406,35 @@ class NRKTVSerieBaseIE(InfoExtractor):
     def _extract_series(self, webpage, display_id, fatal=True):
         config = self._parse_json(
             self._search_regex(
-                r'({.+?})\s*,\s*"[^"]+"\s*\)\s*</script>', webpage, 'config',
-                default='{}' if not fatal else NO_DEFAULT),
+                (r'INITIAL_DATA_*\s*=\s*({.+?})\s*;',
+                 r'({.+?})\s*,\s*"[^"]+"\s*\)\s*</script>'),
+                webpage, 'config', default='{}' if not fatal else NO_DEFAULT),
             display_id, fatal=False)
         if not config:
             return
-        return try_get(config, lambda x: x['series'], dict)
+        return try_get(
+            config,
+            (lambda x: x['initialState']['series'], lambda x: x['series']),
+            dict)
+
+    def _extract_seasons(self, seasons):
+        if not isinstance(seasons, list):
+            return []
+        entries = []
+        for season in seasons:
+            entries.extend(self._extract_episodes(season))
+        return entries
 
     def _extract_episodes(self, season):
-        entries = []
         if not isinstance(season, dict):
-            return entries
-        episodes = season.get('episodes')
-        if not isinstance(episodes, list):
-            return entries
-        for episode in episodes:
+            return []
+        return self._extract_entries(season.get('episodes'))
+
+    def _extract_entries(self, entry_list):
+        if not isinstance(entry_list, list):
+            return []
+        entries = []
+        for episode in entry_list:
             nrk_id = episode.get('prfId')
             if not nrk_id or not isinstance(nrk_id, compat_str):
                 continue
@@ -465,7 +479,7 @@ class NRKTVSeriesIE(NRKTVSerieBaseIE):
     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/serie/(?P<id>[^/]+)'
     _ITEM_RE = r'(?:data-season=["\']|id=["\']season-)(?P<id>\d+)'
     _TESTS = [{
-        # new layout
+        # new layout, seasons
         'url': 'https://tv.nrk.no/serie/backstage',
         'info_dict': {
             'id': 'backstage',
@@ -474,20 +488,21 @@ class NRKTVSeriesIE(NRKTVSerieBaseIE):
         },
         'playlist_mincount': 60,
     }, {
-        # old layout
+        # new layout, instalments
         'url': 'https://tv.nrk.no/serie/groenn-glede',
         'info_dict': {
             'id': 'groenn-glede',
             'title': 'Grønn glede',
             'description': 'md5:7576e92ae7f65da6993cf90ee29e4608',
         },
-        'playlist_mincount': 9,
+        'playlist_mincount': 10,
     }, {
-        'url': 'http://tv.nrksuper.no/serie/labyrint',
+        # old layout
+        'url': 'https://tv.nrksuper.no/serie/labyrint',
         'info_dict': {
             'id': 'labyrint',
             'title': 'Labyrint',
-            'description': 'md5:58afd450974c89e27d5a19212eee7115',
+            'description': 'md5:318b597330fdac5959247c9b69fdb1ec',
         },
         'playlist_mincount': 3,
     }, {
@@ -520,11 +535,11 @@ class NRKTVSeriesIE(NRKTVSerieBaseIE):
             description = try_get(
                 series, lambda x: x['titles']['subtitle'], compat_str)
             entries = []
-            for season in series['seasons']:
-                entries.extend(self._extract_episodes(season))
+            entries.extend(self._extract_seasons(series.get('seasons')))
+            entries.extend(self._extract_entries(series.get('instalments')))
             return self.playlist_result(entries, series_id, title, description)
 
-        # Old layout (e.g. https://tv.nrk.no/serie/groenn-glede)
+        # Old layout (e.g. https://tv.nrksuper.no/serie/labyrint)
         entries = [
             self.url_result(
                 'https://tv.nrk.no/program/Episodes/{series}/{season}'.format(
@@ -536,6 +551,9 @@ class NRKTVSeriesIE(NRKTVSerieBaseIE):
             'seriestitle', webpage,
             'title', default=None) or self._og_search_title(
             webpage, fatal=False)
+        if title:
+            title = self._search_regex(
+                r'NRK (?:Super )?TV\s*[-–]\s*(.+)', title, 'title', default=title)
 
         description = self._html_search_meta(
             'series_description', webpage,
@@ -596,7 +614,7 @@ class NRKPlaylistIE(NRKPlaylistBaseIE):
             'title': 'Rivertonprisen til Karin Fossum',
             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
         },
-        'playlist_count': 5,
+        'playlist_count': 2,
     }]
 
     def _extract_title(self, webpage):